Intelligent Zoning

This application is a continuation of U.S. application Ser. No. 18/432,592, filed on Feb.,, which is continuation of U.S. application Ser. No./,, filed on Nov. 17, 2021, which claims the benefit of priority under 35 U.S.C. § 119 (e) of U.S. Provisional Application Ser. No. 63/117,298, filed on Nov. 23, 2020, and U.S. Provisional Application Ser. No. 63/117,187, filed on Nov. 23, 2020, the disclosures of which are herein incorporated by reference in their entirety.

Disclosed embodiments are related to intelligent zone formation based on vehicle telematics data and related methods of use.

Vehicle location information, such as global positioning system (GPS) coordinates, is typically utilized to provide navigation services. Vehicles may include telemetry hardware devices that track GPS data during vehicle movement. Vehicles may include other devices and sensors that monitor various other vehicle systems and provide vehicle information such as acceleration, speed, fuel level, airbag status, and/or other parameters.

In some embodiments, a method of generating zones with a telematics system includes obtaining telematics data from one or more telematics devices associated with one or more vehicles, the telematics data including GPS data associated with one or more trips taken by the one or more vehicles. The method also includes identifying stops in the telematics data, identifying a cluster of stops using a clustering algorithm, creating a polygon containing the cluster of stops, and outputting the polygon as a zone to a user.

In some embodiments, a telematics system includes at least one processor and at least one storage medium having encoded thereon executable instructions, that when executed by the at least one processor, cause the at least one processor to carry out a method. The method includes obtaining telematics data from one or more telematics devices associated with one or more vehicles, the telematics data including GPS data associated with one or more trips taken by the one or more vehicles. The method also includes identifying stops in the telematics data, identifying a cluster of stops using a clustering algorithm, creating a polygon containing the cluster of stops, and outputting the polygon as a zone to a user.

It should be appreciated that the foregoing concepts, and additional concepts discussed below, may be arranged in any suitable combination, as the present disclosure is not limited in this respect. Further, other advantages and novel features of the present disclosure will become apparent from the following detailed description of various non-limiting embodiments when considered in conjunction with the accompanying figures.

Conventional fleet management systems employ use of telematics for collecting vehicle data for fleet vehicles. Vehicle data may be transmitted from an onboard monitoring device to a remote subsystem, (e.g., server). Fleet management systems may be configurable to monitor and/or process vehicle data for providing fleet information to a fleet manager. In some cases, a fleet management system offers the ability for a user to create zones (e.g., geofences). A zone is a virtual perimeter or a virtual fence around a real-world area of interest. Zones may be used to identify landmarks that can be further used to categorize different areas of interest such as, an office, customer zone, supplier zone, yard zone, warehouse and home location. Creation of such zones is a manual activity. For example, a fleet management system may display an integrated map (e.g., a map that includes global position information, such as Open Street Maps (OSM)) on a user interface (e.g., a display). A fleet manager may manually identify an area of interest on the map and manually draws boundaries on the map to form an enclosed shape to define a zone. A fleet manager may use a user input device, such as a computer mouse, for example, to manually draw the zone on the map.

A fleet manager may create one or more zones for obtaining an understanding of fleet behavior within the zones. Dependent on the shape and location of each zone, the fleet management system may provide vehicle data corresponding to the zones. The fleet management system provides the fleet manager with information describing fleet activity within each zone. Regular monitoring of fleet activity within zones will aid in managing the fleet by the fleet manager. However, the inventors have appreciated that if a zone does not enclose an intended area of interest, information provided by the fleet management system regarding fleet activity within the intended zone will be inaccurate. Exemplary reasons a zone may not properly capture an intended area of interest includes human error, changes to road infrastructure and or parking areas post zone creation, among other reasons.

In view of the above, the inventors have appreciated the benefits of a telematics system and method configured to generate zones automatically based on recorded fleet data. The generated zones may be provided to a user such as a fleet manager who may adopt the generated zones, update existing zones using the generated zones, or obtain zone information without manual creation or manipulation of zones. The telematics system and method may simplify the usage of the telematics system for a user, and/or may reduce the amount of input from the user to obtain actionable information from the recorded fleet data. The inventors have also appreciated the benefits of a telematics system configured to provide vehicle information based on the one or more zones generated by the telematics system. The vehicle information may be provided to a user such as a fleet manager to inform management decisions, for example.

In some embodiments, a telematics system may include at least one processor and at least one storage medium (e.g., non-transitory storage) having encoded thereon executable instructions, that when executed by the at least one processor, cause the at least one processor to carry out a method. In some embodiments, the at least one processor may be a component of a server or multiple servers. The at least one processor may be configured to receive telematics data from a plurality of telematics devices disposed on a plurality of vehicles. The telematics devices may transmit the telematics data to the at least one processor via one or more wired or wireless networks. The telematics data may be stored on the at least one storage medium. In some embodiments, at least one processor may be configured to provide the telematics data or one or more users. For example, the at least one processor may respond to a user request for the telematics data, or a portion thereof. The user may employ the telematics data to inform fleet management decisions. In some embodiments, the at least one processor may be configured to transmit telematics data or a portion thereof to one or more users. For example, the at least one processor may transmit information in a variety of formats in predetermined intervals to one or more users. In some embodiments, the at least one processor may be configured to execute a method of generating one or more zones based on the telematics data from the plurality of vehicles. The one or more zones may be generated based on a request received from a user (e.g., as user input) or may be generated automatically based on the telematics data. In some embodiments, the one or more zones may be generated based on a predetermined interval (e.g., daily, bi-weekly, weekly, bi-monthly, monthly, quarterly, semi-annually, annually, etc.). Exemplary methods that may be performed by at least one processor of a telematics system are described further below.

In some embodiments, a method of generating a zone with a telematics system includes obtaining telematics data from one or more telematics devices associated with one or more vehicles, the telematics data including GPS data associated with one or more trips taken by the one or more vehicles. In some embodiments, the telematics data may be obtained from a plurality of telematics devices each associated with a vehicle. The plurality of telematics devices may transmit the telematics data to at least one processor of the telematics system, for example, by way of one or more networks (e.g., local area network, wide area network, etc.) In some embodiments, the telematics data may be obtained from a storage medium. For example, the telematics data may be historical telematics data stored on the storage medium. The method may also include identifying stops in the telematics data. Stops may correspond to times when a vehicle associated with a telematics device is not moving. For example, stops may include, but are not limited to, idling, vehicle ignition off state, and being positioned off of a roadway. The method may also include identifying a cluster of stops using a clustering algorithm. For example, in some embodiments, a density based clustering algorithm such as DBSCAN may be employed to identify the cluster of stops. The method may also include creating a polygon containing the cluster of stops. For example, the polygon may be formed by the outermost stops within the identifying cluster. In some embodiments, the polygon may have a predetermined shape (e.g., rectangle, pentagon, hexagon, etc.). In such embodiments, the polygon may be sized to contain the stops of the identified cluster. In some embodiments, the method may also optionally include outputting the polygon as a zone to a user. For example, in some embodiments the polygon may be displayed on a base map to a user (e.g., on a graphical user interface of a display). In some embodiments, the method may include storing the polygon as a zone in a storage medium. The stored polygons may be later retrieved by a user (e.g., based on a user request) or sent to a user.

In some embodiments, a method of generating zones with a telematics system may include obtaining telematics data from one or more telematics devices associated with one or more vehicles, the telematics data including GPS data associated with one or more trips taken by the one or more vehicles, and identifying stops in the telematics data, as discussed above. The method may also include identifying a first cluster of stops using a clustering algorithm and creating a first polygon containing the first cluster of stops. The method may also include identifying a second cluster of stops using the clustering algorithm and creating a second polygon containing the second cluster of stops. In some embodiments, the first polygon and the second polygon may be output to a user as a first zone and a second zone, respectively. In some embodiments, a plurality of clusters may be identified by the clustering algorithm, and a corresponding polygon may be created. Accordingly, any suitable number of clusters and polygons may be created by methods according to exemplary embodiments described herein, as the present disclosure is not so limited.

The inventors have also appreciated that in some cases it may be beneficial to merge polygons (e.g., zone) if the polygons exceed one or more similarity thresholds. In some embodiments, if a first polygon and a second polygon are within a threshold distance of one another, the method may include merging the first polygon and second polygon into a merged polygon corresponding to a merged zone. The merged polygon may be output to a user as a merged zone (e.g., at a graphical user interface). In some embodiments, the threshold distance may be less than or equal to 100 m, 50 m, 25 m, 10 m, and/or any other suitable distance. In some embodiments, the threshold distance may be negative, such that the first polygon and second polygon are not merged unless the first polygon and second polygon overlap. In some such embodiments, the threshold distance may be less than or equal to −1 m, −5 m, −10 m, −25 m and/or any other appropriate distance. In some embodiments, a first and second polygon may be merged based on the presence of overlap in terms of percentage area (e.g., a threshold percentage of overlapping area between the first polygon and second polygon is exceeded). In such embodiments, the threshold percentage of overlapping area to merge multiple polygons may be greater than or equal to 1%, 5%, 15%, 20%, 25%, and/or any other suitable percentage.

The inventors have also appreciated that it may be desirable to provide a user vehicle information associated with one or more zones generated according to exemplary methods described herein. In some embodiments, the vehicle information may be provided to a user based on one or more parameters received by a user (e.g., as user input). The one or more parameters may be categories of vehicle information recorded by a telematics device and associated with a stop within an identified cluster. In some embodiments, the one or more one or more parameters include at least one selected from a group of dwell time (e.g., time the vehicle spent at the stop), number of entries into a zone, number of exits, entry points to the zone, exit points from the zone, and vehicle types. In some embodiments, a telematics system may provide the vehicle information to a user automatically (e.g., at a graphical user interface). Of course, any suitable vehicle information associated with one or more zones may be provided to a user on request or automatically, as the present disclosure is not so limited.

In addition to the above, the inventors have appreciated that a telematics data may be large, such that identifying stops and/or identifying one or more clusters of stops may be computationally difficult. For example, some fleets may include well over 100,000 vehicles which may make multiple stops daily. Accordingly, the number of stops stored in at least one storage medium may number in the millions or billions for some fleets, in addition to the other telematics vehicle information associated with vehicle trips. Accordingly, the inventors have appreciated the benefits of parallelizing performance of a method of generating zones according to exemplary embodiments herein for some telematics data sets. In particular, the inventors have appreciated that identifying stops based on more filter parameters input by a user may reduce the size of a data set and correspondingly decrease computational load. Additionally, the inventors have appreciated the benefits of identifying one or more clusters within multiple discrete geographical spaces and combining the geographical spaces following the cluster identification. By identifying one or more clusters within geographical spaces, the search space for a clustering algorithm may be reduced compared with applying a clustering algorithm to an entire data set. Such reduced search spaces may be employed to reduce computational difficulty. Accordingly, methods described herein employing such reduced search spaces may be applicable to many different fleet sizes that have correspondingly different data set sizes.

In some embodiments, stops in telematics data may be identified based on one or more filter parameters received from a user as user input. The one or more parameters may include at least one selected from a group of vehicle type (e.g., delivery vehicle, personal vehicle, light truck, tractor trailer, construction vehicle, etc.), time of day (e.g., 9 AM-5 PM, 5 PM-9 AM, etc.), month range (e.g., November to January, May to July, etc.), date range (e.g., Date 1 to Date 2, last 6 months, etc.), stop time (e.g., greater than 5 minutes, greater than 15 minutes, greater than 30 minutes, greater than 1 hour, etc.), and geography (e.g., country, region, state, county, city, warehouse, factory, store, etc.). Of course, any suitable filter parameters may be employed, as the present disclosure is not so limited. These one or more filter parameters may remove stops from a subsequent clustering identification process. Accordingly, a clustering algorithm may have less data points to process, thereby reducing computation load when compared with identifying clusters on an entire telematics data set.

In some embodiments, one or more filter parameters may be received from a user as user input, and the one or more filter parameters may be based on areas relevant to the user. That is, a user may select the one or more filter parameters such that one or more zones generated according to exemplary embodiments herein are relevant to problems faced by the user. For example, a fleet manager may be only interested in activity of the fleet from 9 AM-PM and may accordingly select this time of day as a filter. Accordingly, zone(s) generated based on this selected filter parameter will be based only on stops made in this time period, such that the generated zones are relevant to the interest of the user. In some embodiments, a user may select multiple filter parameters to further narrow the generated zones to the relevant interests of the user. For example, a fleet manager may be only interested in tractor trailer stops during a delivery window of 12 PM-5 PM. According to this example, the fleet manager may input tractor trailer as a vehicle type filter parameter and a time-of-day filter parameter matching the desired time window. The zones generated based on these selected filter parameters will be relevant to the fleet manager's interest areas. In this manner, one or more filter parameters may be combined in any order and number so that a user may generate relevant zones of interest. As discussed above, the selection of filter parameters may also have the added benefit of reducing a search space for a clustering algorithm.

In some embodiments, identifying a cluster of stops includes assigning a geographical space having a geographical boundary for the clustering algorithm. Assigning the geographical space may exclude stops that are disposed outside of the geographical boundary. Accordingly, a clustering algorithm may identify the cluster based only on the stops within the geographical space, thus reducing the computational load of identifying clusters across an entire telematics data set. In some embodiments, the geographical boundary may be based on one or more geopolitical boundaries. For example, in some embodiments, a geographical boundary includes at least one selected from a group of a national border, a county border, and a postal code border. For example, in an embodiments where telematics data includes stops disposed in a first country and a second country, the geographical boundary may be a national border. According to this example, one or more clusters within the first country may be identified, and one or more clusters within the second country may be identified. In some embodiments, this identification in each country may occur in parallel, such that the speed at which clusters are identified in both countries is accelerated. Similarly, in another example, telematics data may include stops disposed in a first state and a second state, and the geographical boundary may be a state border. According to this example, one or more clusters within the first state may be identified, and one or more clusters within the second state may be identified separately (e.g., in parallel). Of course, any suitable geographic space and geographical boundary may be employed, as the present disclosure is not so limited.

According to some exemplary embodiments described herein, cluster identification may be based on an assigned geographical boundary, where the geographical boundary excludes stops disposed outside of a geographical space. In some cases, there may be a plurality of stops that would otherwise form a cluster disposed on opposite sides of the geographical boundary. As the geographical boundary splits the plurality of stops, a cluster formed within the geographical boundary, if identified, may not be representative of an appropriately sized zone. Accordingly, the inventors have appreciated a method to capture zones disposed near a geographical boundary imposed during cluster identification. In some embodiments, the method may include establishing a buffer extending a predetermined distance around the geographical boundary. In some embodiments, the predetermined distance may be between 50 m and 200 m, 75 m and 125 m, 100 m and 500 m, and/or any other appropriate distance. The method may also include identifying stops within the buffer and identifying a buffer cluster of stops within the buffer using the clustering algorithm. Accordingly, in this manner, the buffer around the geographical boundary may effectively function as a separate geographical space and clusters within the buffer are identified. The method may also include creating a buffer polygon containing the identified cluster of stops within the buffer. In some embodiments, the buffer polygon may be output to a user as a zone. In some embodiments, the method may include determining whether the buffer polygon overlaps with any polygons in the geographical space(s) having the geographical boundary. Based on the determination of whether the buffer polygon overlaps with any other polygons, the buffer polygon may be merged with the overlapping polygon(s). The merged polygon may be output to a user as a merged zone. In this manner, a telematics system and method according to exemplary embodiments herein may identifying clusters within multiple geographical spaces in parallel, and capture zones disposed on or near boundaries of the geographical spaces.

It should be noted that the above methods for reducing computational load to identify one or more clusters of stops from telematics data may be employed alone or in combination. For example, in some embodiments one or more filter parameters may be employed to reduce the number of identified stops within a telematics data set in combination with employing assigned geographical spaces to identify clusters of stops in parallel. In other embodiments, one or more filter parameters may be employed without the use of geographical spaces. For example, if the total number of identified stops based on one or more filter parameters is below a threshold number of stops (e.g., less than 10,000, less than 1,000, etc.), no geographical spaces may be assigned. In other embodiments, a geographical space may be assigned with no filter parameters. Of course, in still other embodiments, stops and clusters of an entire telematics data set may be identified with no filter parameters or assigned geographical spaces, as the present disclosure is not so limited.

The inventors have also appreciated the benefits of a telematics system that may automatically update zones or suggest updated zones for a user. In some cases, fleet deployments, delivery or pickup locations, fueling locations, rest stops, and/or road networks may change or be redeveloped. For example, a particular location associated with stops may close completely, may have a parking area reorganized or enlarged, or may move (e.g., loading zones moved to an opposite side of a building). Accordingly, existing zones may not capture new stops that fall outside of the existing zone due to these changes. In some embodiments, one or more polygons associated with one or more zones may be updated (e.g., at a predetermined interval or on user request). In some embodiments, updating one or more zones includes performing a method of generating zones according to exemplary embodiments described herein. For example, the method of generating zones may include identifying stops in telematics data, identifying a cluster of stops using a clustering algorithm, and creating a polygon containing the cluster of stops. The created polygon may be compared with an existing polygon corresponding to an existing zone. In some embodiments, if the change in area of the existing zone compared to the newly generated zone is greater than a modification threshold, the existing polygon may be merged with the newly created polygon to form an updated polygon corresponding to an updated zone. For example, if the difference in area between the existing polygon and the newly created polygon is greater than or equal to 15%, 20%, 25%, 30%, or another suitable percentage, the existing polygon and newly created polygon may be merged. In some embodiments, if the change in area of the existing zone compared to the newly generated zone is greater than a modification threshold, the existing polygon may be replaced by the newly created polygon. The existing polygon would then be deleted, and the newly created polygon would be used until the next update.

According to exemplary embodiments described herein, a telematics device may be configured to obtain information from one or more onboard sensors and/or one or more sensors or a vehicle control system of a vehicle. In some embodiments, a telematics device may communicate with a vehicle's OBD II port for gathering vehicle operation information therefrom, and may include other sensors/detectors, such as a GPS tracker, for detecting a vehicle's GPS location. Specific and non-limiting examples of vehicle data that may be obtained by a telematics device and transmitted to at least one processor includes onboard monitoring device ID data, position data (e.g., GPS position data), vehicle speed data, ignition state data, (e.g., whether vehicle ignition is ON or OFF), and date and time data indicative of a date and time vehicle operating conditions were logged by the telematics device.

According to exemplary embodiments described herein, an onboard telematics device intermittently transmits vehicle data over a network which is received and stored by, for example, a remote server for future analysis of a single vehicle or fleet performance. A single fleet may comprise any number of vehicles each of which may include a telematics monitoring device. For instance, a fleet may include 100,000 vehicles. In this example, large volumes of vehicle data (e.g., terabytes, petabytes, exabytes, etc.) may be transmitted to, and stored by, the remote server. Of course, methods and systems according to exemplary embodiments described herein may be employed with a fleet of any size, as the present disclosure is not so limited. For example, a fleet may include between 1 and 100 vehicles, 100 and 500 vehicles, 500 and 1000 vehicles, 1000 and 10,000 vehicles, greater than 50,000 vehicles, and/or any other number of vehicles.

According to exemplary embodiments described herein, a user may receive information and/or outputs from a telematics system using any suitable arrangement. In some embodiments, a user may receive information and/or outputs from the telematics system via the internet or one or more wide area networks. For example, the telematics system may include at least one processor of at least one remote server, which may transmit information over the internet to a user. The user may access the information and/or outputs via a web portal, application programming interface (API), data stream, remote connection, or any other suitable arrangement. In some embodiments, the user may receive the information and/or outputs at a graphical user interface of a device such as a smartphone, personal computer, tablet, or other device. In some embodiments, a user may send inputs to the telematics system via the internet or one or more wide area networks. For example, the telematics system may include at least one processor of at least one remote server, which may receive information over the internet from a user. The user may submit the user input via a web portal, application programming interface (API), data stream, remote connection, or any other suitable arrangement. In some embodiments, the user may send the user input at a graphical user interface of a device such as a smartphone, personal computer, tablet, or other device.

It should be noted that while exemplary embodiments described herein are described with reference to one or at least one processor, any suitable number of processors may be employed. For example, the at least one processor may be a plurality of processors of a plurality of remote services. Likewise, it should be noted that while exemplary embodiments described herein are described with reference to one or at least one storage medium, any suitable number of storage mediums may be employed, including a plurality of storage mediums. In some embodiments, a storage medium may be a non-transitory storage medium.

While in some embodiments described here a density based clustering algorithm (e.g., DBSCAN) is employed, any suitable clustering algorithm may be employed, as the present disclosure is not so limited. In some embodiments, a clustering algorithm may be one selected from a group of k-means, affinity propagation, mean shift, spectral clustering, and agglomerative clustering.

Turning to the figures, specific non-limiting embodiments are described in further detail. It should be understood that the various systems, components, features, and methods described relative to these embodiments may be used either individually and/or in any desired combination as the disclosure is not limited to only the specific embodiments described herein.

is a schematic representation of one embodiment of a telematics systemfor performing various processes described herein. The telematics systemis configured to receive information from multiple vehiclessuch as vehiclesdisposed on a road. Each vehiclemay be associated with a telematics device(e.g., vehicleis associated with telematics device). As used herein, the term “telematics device” may refer to any device configured to monitor operation of a vehicle to which it is connected and to communicate data to a remote site. It should be appreciated that the term “telematics device” may refer to a device that is removably installed in and/or integral to a vehicle, as the present disclosure is not limited in this regard.

Each telematics deviceis configured to collect (or otherwise receive) telematics data and to transmit the telematics data through a communication networkto one or more destinations. Such destinations may include a serverincluding at least one processor. As used herein, the term “telematics data” may refer to any data collected, received, analyzed, processed, communicated, or transmitted by a telematics device. While for case of illustration, one serveris shown, it should be appreciated that servermay be implemented as one or more servers, including a distributed system of servers that operate together, such as a cloud service. Such server(s)may be implemented as any suitable form of computing hardware, as embodiments are not limited in this respect. The servermay include software such as a zone generation facilitythat carries out the techniques described herein. It should be appreciated that a zone generation facility need not be associated with a server, but rather that a zone generation facility may be executed on any suitable hardware, as the present disclosure is not limited in this respect.

Transmission by the telematics devicesvia the network(s)may include any suitable transmission technique, including communication to a satellite, through a ground-based station, over a cellular network, over a computer network, over the Internet, and/or using any other suitable channel. Accordingly, network(s)may include any suitable one or combination of wired and/or wireless, local-and/or wide-area communication networks, including one or more private or enterprise networks and/or the Internet. In some embodiments, the telematics devicesmay transmit data using a wireless connection to a wireless wide area network (WWAN) such as a cellular network, after which it may be transmitted via one or more other networks (e.g., wired networks) to a destination such as a server. In some embodiments, a telematics devicestreams data (e.g., contemporaneously with the data being generated and/or received by the telematics device, or in real time) to the servervia the network(s).

The telematics devicesofmay include suitable hardware and/or software configured to collect, sense, receive, process, store, and/or transmit any appropriate telematics data associated with a vehicle. A telematics devicemay be integrated into the vehicle or may be removably connected to a vehicle, such as through a diagnostic port (e.g., an on-board diagnostics (OBD) or OBD-II port). In some embodiments, telematics devicemay include a hand-held device, which may include a mobile device such as a cellular telephone or smartphone. The telematics devicemay communicate with one or more components of the vehicleor otherwise receive from the vehicletelematics data related to the vehicle. The telematics devicemay be configured to connect to a vehicle through an OBD II port or a CAN bus port, or may be integrated into a vehicle's central control system. The telematics devicemay then transmit the telematics data from the deviceand vehicle, to a destination remote from the vehicle. It should be appreciated that the term “telematics device” may refer to hardware and/or software, as the present disclosure is not limited in this regard.

In some embodiments, vehiclemay be configured to collect and transmit information collected by sensors disposed in the vehiclesor otherwise collected by or relating to components of the vehicles, such as through telematics devicesinstalled in the vehicles. The information collected and transmitted may include telematics data for the vehicle. The telematics data collected (e.g., received) by telematics devices in a number of vehicles may be transmitted to a remote site for analysis, such as by processes running on one or more servers. The telematics data that is collected, transmitted, and analyzed may include data generated by a number of different sensors of a vehicle, such as ignition sensors, ambient temperature sensors, fuel sensors, speed sensors, and so on.

In some embodiments, vehiclesmay transmit telematics data, which may include location data (e.g., GPS data) and/or time data. The telematics data may be analyzed to identify locations and/or times where a vehicle was stopped. As such, stops included in telematics data may be associated with a zone of related stops. As also described above, telematics data may include other information in addition to location and/or time data (e.g., ignition data regarding an off or on state of a vehicle). Such telematics data may be collected from many different vehicles. Over time, a large quantity of data may be collected from many different vehicles.

As used herein, telematics data may include data relating to a vehicleor operation of the vehicle. Telematics data may be associated with a telematics devicethat is installed in or integrated with a vehicle. The telematics devicemay receive and/or transmit the telematics data. While telematics data may be associated with a telematics deviceconnected to a vehicle, telematics data may additionally or alternatively be associated with one or more other devices. For example, telematics data relating to a vehicle or operation of a vehicle may be collected, received, and/or transmitted using an application on a smartphone. For example, telematics data related to vehicle location may be associated with location information from a smartphone. As another example, acceleration of a vehicle may be ascertained using data from one or more accelerometers of a smartphone.

In some embodiments, a stop may represent a period of time in which the vehicle is not moving. In some embodiments, a predetermined configurable time period of idling (e.g., 20 seconds or any other configurable time period) may be allowed for and not considered as a stop. Such an arrangement may eliminate traffic or other road conditions as qualifying as a stop. In some embodiments, an ignition off may indicate a stop.

depicts a block diagram of one embodiment of a telematics systemfor generating zones and zone information. One or more components may be, for example, executed as software by at least one processor. As shown in, the systemmay include a zone generatorthat generates zones for vehicle stops based on inputs. In particular, the zone generator generates zones based at least partly on map dataobtained from one or more map information providers (e.g., Open Street Maps). The map datamay include geographic information of a road network associated with a city, a county, and/or a country, as well as geographical boundaries. The zone generator may also generate zones based on telematics data. The telematics data may include stop information of one or more vehicles, as well as other information related to the one or more vehicles. Optionally, the zone generator may operate based on input of filter parameter(s). The filter parameters maybe received as input by a user and may be employed to filter out various stops from being associated with a zone. For example, a filter parameter may include vehicle type, such that only stops associated with a particular type of vehicle (or multiple types) is employed by the zone generator. As shown in, based on the zones output by the zone generator, zone informationassociated with the zones may be provided to a user. The zone information may include telematics data related to identified stops within the zones output by the zone generator. The methods described herein may be implemented by the system of, in some embodiments.

is a schematic of one embodiment of telematics data employed to generate zones.is representative of a mapdepicting a geographical location with telematics data overlaid the map. In particular,depicts a geographical location with vehicle stopsshow by dots. As shown in the, the map includes a first roadand a second roadThe map also includes a parking lotassociated with a distribution building. The distribution building may be a warehouse, factory, or another building that is an origin or destination for vehicles within a fleet. Each dot inrepresents a vehicle stop. As shown in, the vehicle stops may be spread across the various areas of the map, including the first roadsecond roadand the parking lot. The stops may be collected over time from a plurality of vehicles. The stops may be included as a part of telematics data sent to a remote server including at least one processor by a telematics device associated with each vehicle. In some embodiments, the view shown inmay be representative of a display or a portion of a display on a graphical user interface

In some embodiments as discussed above, the stopsmay be filtered based on one or more filter parameters. The one or more filter parameters may be based on user input. For example, a filter parameter may be a vehicle type. Accordingly, the stopsshown inmay all be associated with the selected vehicle type, and no other vehicle types. As another example, a filter parameter may be time of day. Accordingly, the stopsshown inmay have all occurred within the selected time of day. In some embodiments, multiple filter parameters may be employed to filter stops. For example, both vehicle type and time of day may be employed to filter stop displayed on the map. Of course, any suitable parameters may be employed to filter stops, including vehicle type, time of day, month range, date range, stop time, and geography, as the present disclosure is not so limited.

According to the embodiment of, the mapmay include a user created zone. In this example, the fleet manager is interested in fleet activity within the parking lotassociated with the building. For example, the fleet manager may like to know the average amount of time a fleet vehicle waits in the parking lotwhen making a delivery to the building. Accordingly, the user may employ provide user input (e.g., at a graphical user interface of a fleet management system) and draw boundaries on the mapto create the zone, as shown. However, as shown in, such user created zonemay not reflect the identified stops collected from vehicles. Accordingly, the user defined zonedoes not capture many of the vehicle stopsthat are still disposed in the parking lotof the building. Once the zonehas been created by the user, the fleet management system may provide fleet activity information within that zone on a regular basis to the fleet manager for monitoring fleet activity in that zone. However, in the present example, fleet activity information based on the shape and location of zoneprovided by the fleet management system will not accurately reflect fleet activity in the intended area of interest, namely parking lot. Since the fleet manager receives incorrect information regarding fleet activity in zone, proper management of the fleet may be impeded.

depicts one embodiment of a method for generating zones automatically based on telematics data associated with vehicles, and in particular generating zones based on recorded vehicle stops. In block, a first step may include configuring one or more filters for filtering vehicle data by a fleet management system for automatically creating one or more zones. Specific and non-limiting examples of configurable filters include vehicle type, date range, time of day, and stop duration. The vehicle type specifies the type of fleet vehicle for consideration. For example, a vehicle type of heavy-duty truck may be selected, since passenger cars in a fleet may not make deliveries. The date range specifies the date range of stops for consideration. For example, only stops made in the last six months may be included. Time of day specifies a window of time during the day for consideration. For example, fleet activity at all times of the day may be selected (e.g., between 00:00-23:59). As an alternative example, fleet activity at a limited time of the day such as between 09:00-17:00 may be selected. The stop duration specifies a minimum length of time a vehicle must be stopped to be considered a stop. For example, the stop duration filter may be set to a minimum stop duration of 5 min (300 sec). In this example, any stop less than 5 min by a fleet vehicle will be ignored by the fleet for the purposes of zone generation.

According to the method of, in block, a second step may include filtering GPS points corresponding to stops according to the filter configuration set in block. Filtering the GPS points in stepmay include identifying a plurality of stops that meet the filter parameters established in block. In block, a third step includes implementing a clustering technique for clustering the stops identified in block. In other words, in blockhigh density regions of stops are identified as one or more clusters. As discussed previously, any suitable clustering algorithm may be employed to identify the one or more clusters. For example, a Density Based Spatial Clustering technique may be employed for clustering stops and filtering out noisy points (e.g., outliers). In some embodiment, such a technique utilizes pairwise distance between stops for determining stop density. The third step may also include removing, outlier stops that do not fall into a cluster in order to provide a most accurate shape and location of a zone/zones that are ultimately generated based on the identified one or more clusters. As shown in block, stops associated with clusters may be stored (e.g., on a non-transitory storage medium) for later use.

As shown in, in blocka fourth step includes creating a polygon for stops that are within the same cluster identified in stepas described above. In some embodiments, forming a polygon includes forming a convex hull around each cluster of stops. The polygon may ultimately determine a zone, and in this sense the polygon is a representation of a generated zone. In block, vehicle data corresponding to stops (e.g., locations) within a zone may be processed to provide one or more of the following descriptive statistics for each zone. Some examples of descriptive statistics include stop count in the conferred time period, distinct vehicle count in the configured time period, stop distribution by month, average stop duration, 15 percentile stop duration, 50 percentile stop duration, 85 percentile stop duration, average driving duration before stopping in the zone, and average idling time in the zone. The metrics may assist a user such as a fleet manager in developing some context around the newly generated zone(s). In block, the zones and associated information may be output to a user (e.g., at a graphical user interface).are described further below which depict several examples of implementations of methods according to exemplary embodiments described herein, including the method of.

depict steps of one embodiment of a method of generating zones employing the telematics data of. As show in, method according to that ofhas been performed on the vehicle stops. As a result, a first polygoncorresponding to a first zone and a second polygonare generated based on two clusters of vehicle stops identified in the parking lot. Vehicle stops which do not correspond to either of the two polygons are discarded, shown by the stars. As shown in, the starts are located in positions that do not correspond to the dense clusters defining the first polygonand the second polygonFor example, some of the outliers are disposed on the roadsor more adjacent the building. According to the embodiment of, the first polygonand second polygonmay contain each stop associated with a cluster.

According to the embodiment of, a clustering algorithm generated two discrete polygonsHowever, the inventors have appreciated that zones within a predetermined distance from one another, or overlapping zones, may represent a singular zone encompassing an area like a parking lot with a geometry that creates adjacent clusters. Accordingly, in some embodiments, a first polygonand a second polygonmay be merged with one another if certain conditions are met. In the embodiment of, if a distance D between the first polygonand second polygonis below a threshold distance, the first polygon and second polygon are merged. For example, in some embodiments, if the distance D is less than 25 m, the first polygon and second polygon may be merged. Of course, any suitable threshold distance may be employed, as the present disclosure is not so limited.depicts the results of a merged polygonencompassing both the first polygon and second polygon of.

depict steps of another embodiment of a method of generating zones. According to the embodiment of, a regional mapis shown including a plurality of roadways. The display shown inmay correspond to that shown on a graphical user interface to a user. As shown by the dots overlaid on the map, vehicle stopsare identified from telematics data. A method of generating zones (e.g., as discussed with reference to), may be employed that applies a clustering algorithm to the vehicle stops. Once one or more clusters are identified, polygons corresponding to zones may be generated and output to a user.depicts the result of such a method. As shown in, three polygonswere generated corresponding to three zones. Accordingly, methods and systems herein may generate multiple zones across an entire geographical region and provide those zones to a user such as a fleet manager. Vehicle information associated with each zone may also be provided to the user.

depict steps of another embodiment of a method of generating zones. As discussed previously, in some embodiments a clustering algorithm may be applied to vehicle stops disposed in a geographical space including a geographical boundary. Such an arrangement may allow multiple clustering algorithms one multiple geographical spaces to be executed in parallel. Accordingly, clustering algorithms may be applied to adjacent geographical spaces separately, and resulting zones may be separated by geographical boundaries, which is shown in the exemplary embodiment of. As shown in, a geographical boundary B divides the vehicle stops, which are disposed on either side of the boundary. A clustering algorithm may be performed separately on vehicle stops on either side of the boundary. Accordingly, as shown in, the resulting polygons will be different than those output in the embodiment of, for example. In particular, three polygonsare generated that are associated with the lower geographical space, and two polygonsare generated that are associated with an upper geographical space. A first polygonof the lower geographical spacewould be combined with a first polygonof the upper geographical region if the boundary limitation was not present. Accordingly, in some embodiments a method may compensate for this divided zone result, as will be discussed with reference to.

depict steps of another embodiment of a method of generating zones including compensating for a division of clusters by a geographical boundary B. As shown in, the boundary B divides a plurality of vehicle stops. Accordingly, where a clustering algorithm is executed separately on the upper geographical space and the lower geographical space, multiple zones may be formed where there should be one zone. Accordingly, in some embodiments as shown in, a buffer may be established around the boundary B, as shown inas lines Aand A. The buffer may extend a predetermined distance from the boundary (e.g., 100 m, 50 m, etc.). In some embodiments, a method of generating zones may include treatment of the buffer as its own geographical space. That is, in some embodiments, vehicle stops within the buffer may be identified, one or more buffer clusters of stops may be identified using a clustering algorithm, and one or more buffer polygons may be created corresponding to the one or more identified buffer clusters. The results of such a method are shown in. As shown in, a buffer polygonhas been created based on the identified cluster of stops that straddle the geographical boundary B., depicts the results of a method according to, where a polygonof the lower geographical space and a polygonof the upper geographical space are created based on the separate execution of a clustering algorithm in each geographical space. In some embodiments, a method may include determining whether the buffer polygon and the polygonor the polygonoverlap. If the buffer polygon and any of the polygons,do overlap, the buffer polygon may be merged with the overlapping polygons,. Accordingly, as shown in, the resulting merged polygonencompassed all vehicle stops within the cluster on both sides of the geographical boundary. Thus, the method shown inmay be employed to ensure zones generated by methods and systems according to exemplary embodiments described herein do not limit appropriate clusters by introduction of geographical boundaries during computation.

depicts another embodiment of a method for generating zones. As an example to illustrate the method of, an exemplary fleet operates 24/7 and comprises 200,000 vehicles, including 100,000 heavy duty trucks (HDTs) 90,000 light duty trucks (LDTs), and 10,000 passenger cars. According to this example, a user wishes to obtain information relating to stops made by the fleet vehicles in the last 6 months. In block, a first step includes configuring one or more filter parameters for filtering vehicle data to generate one or more zones. As discussed above, the filter parameters may include one or more parameters which may be input by the user. In the present example, a user may apply a vehicle type filter parameter for the HDTs and LDTs, thereby filtering the stops of the passenger cars. In the present example, a user may apply a time filter to filter stops not made in the last six months. Of course, any suitable filter parameters may be selected, as the present disclosure is not so limited. In block, a second step includes selecting vehicle data corresponding to filter configurations of blockfor identifying vehicle stops based on telematics data. According to the example, 100 million stops may be obtained after block.

As shown in, in blocka third step of implementing a clustering technique for identifying high density regions of stops begins. In the present example, 100 million stops are to be clustered prior to creating zones. In some cases, such a large number of stops introduces a scaling challenge with respect to applying the same clustering technique to both small fleets and large fleets (e.g., the exemplary fleet). The inventors have appreciated that a limitation of density based clustering algorithms is that such algorithms compute pairwise distances to determine the density. For a smaller fleet, processing time and/or power is not an issue when using such an algorithm for clustering stops. However, applying such an algorithm to a number of stops associated with a large fleet, such as 100 million stops, may present a bottleneck in the zone creation process. However, the inventors have appreciated that by reducing the search space for the spatial clustering algorithm, the challenge may be alleviated and enable the use of the same approach for different fleet sizes. Accordingly, as shown in block, prior to implementing a clustering technique for identifying high density regions of stops, the method includes partitioning the stops according to a specified criteria. Once partitioned, in blockthe method includes clustering each partition of stops in parallel. For example, in some embodiments, stops may be partitioned according to geography. For instance, stops may be partitioned according to administrative or geographical boundaries for governing different regions, e.g., country, state/province, city, division, county, municipality postal code, and/or any other suitable boundary. As another example, in some embodiments, stops may be partitioned by one or more filter parameters, such as vehicle type. For example, a clustering algorithm may be executed on stops associated with heavy duty trucks in parallel with executing a clustering algorithm on stops associated with light duty trucks. In block, clusters identified as a result of blockmay be stored in a storage medium.

As discussed previously, partitioning stops helps to scale the computation intensive clustering process, but it also presents potential challenges related to splitting a zone disposed on an administrative boundary. Accordingly, as shown in, In blockstops that are disposed within a buffer region around a geographical boundary are identified. The buffer region may extend a predetermined distance from the geographical boundary. In block, one or more clusters of the buffer stops may be identified (e.g., by a clustering algorithm). The identified buffer clusters may be stored in a storage medium as shown in block. As shown in, the clusters identified and stored in blockand the buffer clusters identified and stored in blockare combined in block. In block, it is determined if any of the combined clusters overlap. In some embodiments, if any clusters overlap, the overlapping clusters are merged. In some embodiments, if any clusters are within a threshold distance from one another, the clusters within the threshold distance of one another are merged. In block, a fourth step of the method is creating a polygon for each of the identified clusters following the merging process of block. The created polygons may each correspond to a zone.